NanoString Technologies Inc (NSTG) 2021 Q2 法說會逐字稿

Good day, and thank you for standing by. Welcome to the NanoString's Second quarter 2021 Operating Results. (Operator Instructions) Please be advised that today's conference is being recorded. (Operator Instructions)

Without a further ado, I would like to welcome your first speaker for today, Mr. Doug Farrell, Vice President of Investor Relations. Sir, the floor is yours.

Thank you, operator. Thank you, everyone, for joining us today. On the call with me today is Brad Gray, our President and CEO; as well as Tom Bailey, our CFO. Earlier today, we released our financial results for the second quarter of 2021. During this call, we may make statements that are forward-looking, including statements about financial projections, the impact of the COVID-19 pandemic, future business growth, trends and related factors, prospects for expanding and penetrating our addressable markets, our strategic focus and objectives, and the development status and anticipated success of recently planned product offerings. Forward-looking statements are subject to risks and uncertainties, many of which are beyond our control, including the risks and uncertainties described from time to time in our SEC filings. Our results may differ materially from those projected, and we undertake no obligation to update these forward-looking statements.

Later in the call, Tom will be discussing our financial results and 2021 guidance. We have prepared as a supplement to GAAP financial measures, selected non-GAAP adjusted measures, the calculation of which are described in detail in our press release. Throughout this call, all financial measures will be GAAP, unless otherwise noted. You can find reconciliations of GAAP to non-GAAP, as well as the description, limitations and rationale for using such measures in afternoon's press release, which are also available under the Investor Relations page on our website. I'd like to remind everyone we are participating in the UBS Genomics 2.0 conference next week, as well as the Baird & Morgan Stanley Healthcare conferences in September. We look forward to having the opportunity to speak with many of you there.

Now I'd like to turn the call over to Brad.

Thanks, Doug. Good afternoon, and thank you for joining us today. The momentum of our spatial biology franchise continues to increase each quarter. We continue to extend our lead in translational research. While the rapid adoption of NGS read-out or GeoMx is opening up a vast new market in basic discovery research. Our spatial molecular imager has borrowing interest from existing genomic customers, as well as single cell genomics researchers who are engaging with NanoString for the first time. Meanwhile, our nCounter franchise is humming along, continuing to generate strong instrument placements in life science more than a decade after its introduction.

Our first strategic objective for 2021 is to extend our leading spatial biology through the broad adoption of the GeoMx Digital Spatial Profiler. In the second quarter, we generated a record number of orders for GeoMx DSP instruments, and delivered sequential growth in GeoMx consumable pull-through per system and GeoMx instrument orders grew approximately 50% year-on-year at the top end of our guidance range of 40% to 50% growth. The recent launches of our Human and Mouse Whole Transcriptome Atlas or WTA, propelled annualized consumable pull-through to $94,000 per system during the second quarter, near the top of our previously provided guidance range of $85,000 to $95,000. The impact of WTA on consumable pull-through is pushing our genomics revenue expectations for 2021 towards the top half of our previously guided guidance range, and importantly, increases our estimates for the long-term value of our genomics franchise.

Expanding the use of next-generation sequencing to read out GeoMx experiment is a key growth catalyst from our spatial business. The recent launches of our first Whole Transcriptome products are driving the momentum. These are the first system-wide products that we've offered, and they are appealing to researchers in both translational and discovery research. During the second quarter, approximately 60% of new GeoMx instruments were sold to researchers, intending to use NGS as their primary readout. In addition, many of researchers who initially used their GeoMx systems with nCounter based readout, have now begun to utilize assays with NGS readout as well. So far this year, approximately 1/3 of our GeoMx installed base has ordered either our Whole Transcriptome Atlas or a Cancer Transcriptome Atlas assays. Together, these NGs readout assays accounted for approximately 2/3 of total GeoMx consumable revenue in the second quarter.

We expected the expanded research applications of our NGS enabled assays will drive broad adoption of GeoMx, while the higher revenue for several of these assays will continue to increase consumable pull-through system through the balance of the year. During the second quarter, we expanded our NGS-enabled GeoMx portfolio with the launch of the Mouse Whole Transcriptome Atlas. The Mouse is the most important model system for basic discovery research, and our larger research suggest it may represent up to 30% of the spatial biology camp.

In February, we began offering the Mouse WTA through our Technology Access Program, or TAP. This new offering is generating strong interest from researchers with diverse areas of biology, including immunology, neuroscience, infectious disease and developmental biology. Our Technology Access Program enables researchers to test drive GeoMx, while evaluating its instrument purchase, and provides us with another strong indicator of customer enthusiasm for NGS readout. We generated more than 90 new TAP orders in the second quarter, more than doubling the number of projects over the prior year, with NGS readout using about 90% of the new projects.

While our Whole Transcriptome Atlas is driving us in the basic discovery research, we continue to build on our track record of leadership among translational researchers focused on human disease. NanoString has been serving the translational research community for more than a decade, and we have a strong brand and deep understanding of their unique requirements. We understand that while robust and reliable results from Formalin-Fixed Paraffin-Embedded or FFPE tissue there is a terrible state for any platform. Translational researchers demand much more. They want automated solutions that can drive high sample audience and provide consistent site-to-site performance needed for collaborating across networks. They also want to analyze both RNA and protein on the same platform. GeoMx' Digital Spatial biology platform meets all of these requirements, helping establish its market-leading position and translational research.

To be clear, while other companies are adapting their technology to be FFPE compatible, NanoString's platforms were developed using FFPEs from the outset. Recent product updates from other companies have neither shifted the competitive landscape or impacted our leadership in translational research.

Among other requirements, translational researchers value that GeoMx provide protein assays enabled by a portfolio of several hundred validated antibodies and best-in-class plex. For the researchers focused on protein assays, nCounter readout remains the preferred configuration and represents approximately 40% of GeoMx systems sold during Q2. Half of these systems were about 20% of all GeoMx systems sold during the second quarter, were packaged as GeoMx plus nCounter bundles. In most cases, these bundles are going to translational researchers who are new NanoString customers, further expanding our lead in this important market. Overall, GeoMx continues to provide a strong growth driver for 2021 and are set up for long-term success, transforming our company.

Our second strategic objective for 2021 is to advance the development of our spatial molecular imagery and to seed the market for the commercial launch that's planned for next year. Molecular Imaging is shaping up to be an important new market that we expect will be highly complementary to spatial profile. We're making great progress in developing our Spatial Molecular Imager, or SMI, which remains on track to ship commercial instruments in the second half of 2022. Our SMI is expected to be the best-in-class and distinctive from other imagers in several important ways. First, we expect SMI to lead the class with both the highest plex and the highest sensitivity. Second, we expect SMI to show superior performance in FFPE tissue samples. And third, unlike competing platforms, SMI will launch with both RNA and protein assets. We've developed a Technology Access Program for SMI, and we've been receiving interest from a diverse set of researchers.

To date, about half of the SMI TAP projects have come from existing GeoMx customers, while the other half come from single cell researchers, just beginning to embrace spatial biology. The experiments we are running span many areas of biology, including oncology, immunology and neurology. Some researchers are interested in using spatial molecular imaging to create an atlas of cells within a tissue, while others are looking to characterize interactions between individual cells. Our TAP service is already oversubscribed for the year, and we've been allocating our capacity to high-impact science with key opinion leaders. Researchers continue to reach out with interesting projects, creating a reservoir of demand that will keep us busy well into next year.

Our third strategic objective is to return our nCounter business to the growth dynamics seen prior to the COVID-19 pandemic. Our nCounter franchise remains robust, with our installed base increasing about 14% over the prior year and customers publishing another 300 theory papers during the recent quarter. We achieved a major milestone on our nCounter business in Q2, selling our 1,000 systems. As of midyear, our 2021 instrument revenue almost exactly matches what we generated during the same period in 2019. This signifies a full recovery of instrument placements to pre-pandemic levels, and we expect this trend to continue through the second half of the year.

In Q2, we generated another sequential increase in consumable pull-through to about $57,000 per system annualized, with recovery most notable in North America. We expect consumable pull-through to continue recovering from pandemic lows during the second half. Oncology researchers continue to embrace nCounter. We've recently entered an exciting collaboration with the Parker Institute for Cancer Immunotherapy focused on the molecular characterization of cellular therapies. The objective of this collaboration is defined the characteristics that will make a cell therapy effective, providing a standardized approach to developing CAR-T regimens that may improve patient outcomes across all cancer types. The collaboration will leverage the cell therapy expertise of Parker's extensive network of world-class research centers, and Parker's plans to make the findings publicly available to the scientific community.

In another exciting development, we recently entered a long-term umbrella contract to supply the National Institutes for Health, that's part of its INH program. [CNIA] sponsored thousands of single trials per year. And under this agreement, NIH researchers will now have expanded access to our nCounter platform to assess the immune status of solid tumors for patients in clinical trials. We continue to push nCounter into new areas beyond oncology. In the second quarter, about half of our nCounter systems were sold for applications outside oncology and consumable growth was especially strong in infectious disease and immunology.

During Q2, we launched a new Spin Cell Characterization Panel for the analysis and optimization of cell lines used in the development of novel therapeutics. Our new panel measures 8 essential components within cell biology and provides a standardized assay through evaluating factors that impact the viability in the development and manufacturing process. We hope it will help researchers scale up the stem cell therapy workflow, advancing this promising fields.

To summarize, we're making great progress in our strategic objectives, with nCounter returning to growth, GeoMx reaching new customers through a NGS readout and SMI generating excitement ahead of its 2022 launch.

And now I'd like to turn the call over to Tom to review the details of our operating results.

Thanks, Brad, and thanks all for joining us today. For the second quarter of 2021, product and service revenue was $33.6 million, representing year-over-year growth of 59%. Q2 GeoMx revenue was $11.2 million, up 66% as compared to Q2 of '20, and above the high end of guidance we provided in May. $7.4 million, that's approximately 30 instruments shipped and $3.8 million was from consumable sales. GeoMx consumables pull-through was about $94,000 per installed system in Q2. For nCounter, total revenue for Q2 was also above the high end of our May guidance. Q2 instrument revenue was $4.4 million, representing year-over-year growth of 25%. nCounter consumable revenue was $14.2 million, representing year-over-year growth of 80%. Annualized consumables pull-through was about $57,000 in Q2. Service revenue was about $3.8 million for the quarter, representing 29% year-over-year growth and driven by new GeoMx DSP TAP projects, and increased service contract revenue from our growing instrument installed base.

Turning to margins and expenses. I'll provide results on a non-GAAP or adjusted basis, which removes the impact of stock-based compensation, depreciation and certain onetime items. Please refer to our press release, as well as the exhibits we have posted to our Investor Relations web page for detailed information on how our non-GAAP or adjusted measures are prepared. Q2 adjusted gross margin was 56%, consistent with our annual guidance range at about a 400 basis point improvement compared to Q2 of last year, driven by both the growth in GeoMx DSP revenue and the recovery in income of consumable sales compared to the prior year period.

Adjusted R&D expense was $14.5 million, an increase of 6% year-over-year. R&D was higher compared to the prior year period due primarily to increased personnel and product development costs related to our SMI program. We expect R&D expense will increase through the balance of the year as SMI development continues.

Adjusted SG&A expense was $21.7 million, an increase of 27% year-over-year. The Q2 SG&A expense increase was due primarily to investments made in our special biology related commercial initiatives, including investments to expand our sales force and our service and customer support groups. Adjusted EBITDA loss was $17.3 million, an improvement of 12% as compared to the prior year. We exited the quarter with approximately $398 million of cash, cash equivalents and short-term investments.

Turning to guidance. For Q3, we expect product and service revenue to be in the range of $36 million to $38 million, representing year-over-year growth of 20% to 26%. This range assumes a $24 million to $25 million of nCounter revenue and $12 million to $13 million of GeoMx revenues. Regarding our full year guidance, we are raising our full year guidance range for GeoMx consumables pull-through to $95,000 to $100,000 per system per year, implying annualized pull-through of approximately $100,000 to $105,000 for the second half of 2021, based on strong early consumable utilization and NGS-enabled sites. We are also affirming our previous guidance of 40% to 50% growth in instrument orders as compared to the prior year. As a result, we are updating our full year GeoMx revenue guidance range of $48 million to $50 million, representing annual growth of 38% to 43%.

For nCounter, instrument demand has rebounded to pre pandemic levels globally, and our expectations for nCounter instrument revenue remain unchanged. We are narrowing our full year guidance for consumables pull-through to $60,000 to $63,000 per system per year, implying approximately $63,000 to $66,000 for the second half of 2021, and reflecting a rebound in consumables demand in North America that's been partially offset by a modestly slower pace of pandemic recovery in Europe and Asia. As a result of these updates, we are narrowing our full year nCounter revenue expectation to $95 million to $97 million, representing annual growth of 24% to 27%.

The combined impact of these updates is an updated full year product and service revenue guidance range of $143 million to $147 million, representing annual growth of 28% to 32%, with the midpoint of our total revenue guidance remaining unchanged. We are also affirming our prior full year gross margin, operating expense and adjusted EBITDA loss outlook amounts as provided on our March 1 call. Additionally, as a reminder, we do not expect any material collaboration revenue to be recorded in future periods.

Now I'll turn the call back over to Brad for closing comments.

Thanks, Tom. As we enter the second half, the year receiving up just as we had envisioned. Our new Whole Transcriptome Assays are driving GeoMx adoption and increasing consumable pull-through, enhancing the long-term value of our spatial business. Our deep understanding of customer needs, combined with the unique throughput and multi-element capabilities of GeoMx are extending our leadership in translational research. Scientists across many fields of research are intrigued by our spatial molecular imager and are lining up to perform their first experiments. Meanwhile, our nCounter franchise is still achieving important milestones more than a decade after its introduction. Our momentum is strong, and we are confident in both our near-term growth prospects, as well as the long-term value of our spatial biology franchise.

With that, we'd now like to open the line for your questions.

(Operator Instructions)

Our first question comes from the line of Tycho Peterson from J.P. Morgan.

This is Julia on for Tycho today. So I guess in terms of the current mix, I understand you have 2/3 academic and 1/3 biopharma for GeoMx. How do you see the mix kind of changing going forward with the Whole Transcriptome launch? And how should we think about the relative runway between academic versus biopharma? I guess the former, it's a much more fragmented market, but for the latter, it's maybe more ready to kind of step up into translational studies. So just help us think through the relative runway there. Thanks.

You're right. Our installed base for GeoMx is about 2/3 academic and 1/3 biopharma. Initially, when we launched our Whole Transcriptome Assay, we imagined it would appeal more to the basic discovery researchers who you find predominantly in academic research settings. And so we had projected that over time, the fraction of GeoMx that is going to academic research will actually increase. As I mentioned in my prepared remarks, we're seeing that the NGS readout capabilities of GeoMx, including the Whole Transcriptome Assay are appealing both, the basic discovery researchers and translational researchers. So at this stage, I think we should expect to continue to see an approximately 2/3, 1/3 split for new GeoMx instrument placements between academics and biopharma customers.

Got it. That's helpful. And then on kind of the biopharma and transitional research side, is there any plan or room to kind of increase the GeoMx throughput, so you can actually use it in clinical trials?

The GeoMx throughput is already extremely high relative to all of their spatial biology platforms and is probably sufficient for use in a clinical trial. The novel throughput of a GeoMx system is about 10 samples per day. And there's really very few clinical trials that would ever be enrolling patients at a faster pace than that or a need to screen patients for inclusion in any faster pace than that. So we feel really good about the capability of the GeoMx system to serve the translational research market, including clinical trials. So of course, most of our use of GeoMx in translational research today is the retrospective analysis of FFPE samples that were collected in clinical trials. But we think that the throughput could hypothetically allow the prospective analysis at the enrollment of a clinical trial as well, without any material changes to what our GeoMx system is.

Got it. Very helpful. And then lastly, I know you gave color on the TAP order for GeoMx. Could you share the same for SMI? How many early customers you're working with and any upcoming publications we should be paying attention to?

Yes. So in the past, we've said that our plan for the SMI half for 2021 was to enroll 15 to 20 projects worth, and we are fully subscribed at that level now. But we continue to collect excess demand, as I mentioned in my prepared remarks, those interesting projects are sort of on standby until we are able to expand out that capacity in 2022. We aren't previewing any particular forthcoming publications at this point in time, but look for incremental news flow on SMI in the fall and in the fourth quarter as some of these TAP projects begin to make their way through completion and into appropriate public presentations.

Our next question comes from the line of Daniel Arias from Stifel.

Brad, on the consumable side and the raise on the GeoMx pull-through average, can you maybe just add some color to the elements in play there? I think it sounds like next-gen usage is up nicely. So what are you seeing just in terms of assay mix to your point before, and the impact that, that's having, maybe sample numbers relative to your initial expectations? And then any other -- any other sort of residual comments that you might have on utilization trends that are sort of driving that rate forward?

Yes. I think it's been incredibly exciting and even surprising to us, the broad interest in the Whole Transcriptome Atlas Assay. In the past, you've heard us talk about the Whole Transcriptome Atlas as our gateway assay into just the discovery market. But what you've heard in our prepared remarks is that the entirety of our installed base is intrigued by this assay that is beginning to adopt it. That drives the overall dollar per sample up for our entire installed base and is translating into the second raise in our consumable pull-through guidance that you heard today.

I think it's because most customers are still in their, call it, trial usage phase of the Whole Transcriptome Atlas Assay, I think it's too early to characterize exactly what their steady-state run rate of assays per system per year would be. But clearly, the overall dollar per assay average for our installed base is going up. And that's, of course, a key driver of the long-term value of our spatial biology franchise, which was what makes it so exciting.

Yes. Okay. A couple of quarters, you had said that you thought that if you fast forward into the end of this year, Whole Transcriptome could come to make up more than like half of what's done versus Cancer Transcriptome. Do you still think that makes sense as an assumption?

Yes, I do. I mean I think the NGS readout assays, Cancer Transcriptome Atlas and Whole Transcriptome Atlas, together accounted for 2/3 of our consumable revenue on GeoMx in the second quarter. So the Whole Transcriptome Assay is probably already very close to half of the utilization of our total installed base. And I think that will just grow over the -- I think that will strengthen over the balance of the year.

Okay. Okay. And then just maybe on the instrument side, I mean, you guys have done a really nice job just sort of keeping the GeoMx business steady throughout the pandemic. So certainly, kudos there for sure. I'm wondering when you think we start to sort of see a step-up on orders and installs just in the coming quarters. Obviously, the pull-through is a telling sign on usage or one of them anyway, and that's going up. But it does feel like the enthusiasm around the spatial market just sort of being on the cusp of an inflection, can start to show up in the adoption rate, especially since it feels like the constraint is not just getting into the lab are not as high. So totally appreciative of the fact that things are not 100% out there, I do get that. But is it fair for us to model acceleration in the placement rate going forward from the 30 or so that you've been on for a while?

Yes. And I think -- so first, let's separate orders and revenue recognition because those are really important different dynamics. So through the first half of this year, in both Q1 and Q2, our order rate has gone up approximately 50% year-on-year. So that is -- and that's the best indication of actual sort of real-time demand from the market for GeoMx. So that's growing at a very strong rate, at the top end of our guidance rate for the year. And I think we're maintaining our 40% to 50% growth guidance for the full year on orders. So -- and of course, that will -- because when you look at the comparable for the back half of last year, that will increase orders beyond the 30 or so placements that you've seen from us the past couple of quarters. And given that we're in a one-to-one kind of book-to-bill scenario now, where each order is more or less fulfilled with an instrument shipment, you will begin to see instruments growing beyond 30 in the second half.

That being said, part of your questions seem to imply, Dan, that you were wondering if there was going to be an enormous step change or inflection with respect to instrument revenue. I don't think that's the natural dynamic of this market. We are marching orders up at 50% a year. That's a great cliff. I don't expect that suddenly to spike to a higher rate, if we can sustain that pace of inset order growth through the balance of the year, and sustaining that type of growth of 40% to 50% range even in the next year, I think we're great. We're going to build a tremendously value for spatial franchise, without the need for anything that I would characterize as a particular inflection.

Yes. No, no, I certainly wasn't thinking big step-up. I was just -- I'm going back and I'm looking at the installs per quarter, and I'm saying, okay, you guys actually placed 35 systems in Q3 of last year, which again, is a huge accomplishment given how difficult the operating environment was then. So for those of us that are just trying to understand the velocity or the trajectory of the spatial market, is it fair to assume that if we push into the end of this year or the beginning of next year, you'll start to see that number creep up. That just says, hey, there are more people out there that are looking to do spatial experiments, they're ordering boxes, and you guys are putting more of them in labs than you did, say, a year ago.

Yes. Certainly, that growth is expected in the second half. And then if you -- just to work the math for you, we're maintaining our 40% to 50% guidance range for orders, given where we are in the year, that would imply a range of sort of 30 to 35 orders in the third quarter and 35 to 40 orders in the fourth quarter. And that's I just worked math on growth on the 90 that we did full year 2020, and the 60 or so that's been year-to-date. So that gives you a sense of what's implicit in our guidance stand and if we can keep growing instrument placements in that kind of range, I think we're going to be in a good place.

Our next question comes from the line of Catherine Schulte from Baird.

I guess, first, can you just walk through what you're seeing on the reopening side? Where are different regions in terms of activity levels versus pre-COVID levels? And then what are you guys seeing when it comes to access to customer sites?

Yes. Great question, Catherine. So as we alluded to in our prepared remarks, our best indicator for customer activities and the place where the COVID-19 pandemic impacted us the most is on nCounter consumable pull-through. And we've seen, I'd say, an uneven pandemic recovers geographically. In North America, our business is almost back to the pre-pandemic utilization rate of nCounters that we experienced in 2019. That being said, in EMEA and APAC, we remain substantially below what our 2019 utilization rates were. And we attune back to slower recoveries from the pandemic in those regions. That may or may not be the same as what other companies are seeing, but at least within our installed base, which is focused primarily on cancer research and to a lesser degree of immunology and neurology research, that's what we're seeing.

So importantly, maybe I'll just add, none of that pandemic recovery seems to be impacting the placements of nCounter instruments, which are back on the 2019 pace or the pace of GeoMx instrument orders, which are in line that has been at least the comp end of our guidance range so far this year.

Yes. Got it. And I know you guys have talked about hiring about 100 new reps. How is that hiring process going? Where are you in terms of those hires? And how should we think about a potential back half impact?

Yes. We're about 3/4 away through hiring those 100 reps. We've made good, strong progress, especially in North America, where, of course, hiring is fast and simple. The places we still have hiring underway, it's more in Europe and Asia, where it just takes longer to hire people. But we're pleased with the progress. I think we'll begin to start to see the impact of those reps later this year. It usually takes 6 months for a rep to become effective once they joined, once they are trained. So people who were hired in the second quarter would be beginning to be effective in the fourth. And then, of course, we'll have a full year of their impact on 2022.

Yes. If I could sneak one more in. I am just curious if we could talk on the innovation road map for GeoMx from here. You guys have 3 different nCounter platforms. Are there different iterations of GeoMx that you consider coming out over the next several years?

Yes. For now, we're focusing all of our GeoMx innovation on the assay front and the software front. We believe that the combination of price point, footprint and throughput, that our current GeoMx system has is appealing to really core labs across both the discovery and the translational market segments. And what we want to do is continue to innovate on the consumables that will appeal to a larger number of customers. So obviously, the Whole Transcriptome assays were the single most important consumable that we've introduced, but we're also continuing to expand our protein library and fields beyond cancer. We're continuing to look at custom assay options for GeoMx that would allow us to address biology that takes place outside of a human or a mouse model. Those are some of the consumable innovations.

And then separately, we're learning a lot from our customers on how they want to store, to process and to manipulate the data that's generated in spatial biology. And we're moving towards more sophisticated software tools that will improve the customer experience and hopefully drive even higher utilization of the systems that are out there. So today, we don't have a specific plan with respect to instrument innovation, but certainly, we will update you over time as that roadmap develops.

Our next question comes from the line of Tejas Savant of Morgan Stanley.

This is Edmund on for Tejas. Just wanted to circle back on the recovery trends and your nCounter consumables. I know last quarter, you guys had called out EU as a pocket of weakness. I was wondering if you could provide some additional color on the recovery pace specifically for the EU region. And in terms of your guidance, what's baked into for, I guess, delta variants and worsening COVID impacts?

Yes. So maybe I'll take the second part of that question first. I think our guidance today on nCounter consumables does bake in the current uncertainty with respect to the delta variant. I mean we're watching that in real-time as everyone else is. That's part of the reason that we're no longer projecting that we'd be at the top end of our nCounter previously provided guidance range, because with this uncertainty, I think we can only expect that the recovery will be on pace or a little bit slower than we might have originally managed. But I do not see it getting any worse than as embodied in our updated guidance range today.

With respect to Europe, I'm not really in a position to provide too much in the way of additional color. I mean, I will say it's worth remembering that the areas that are weak -- that are relatively weak in terms of demand, which for us are Europe and Asia, are the same regions where we work through distributors in many cases. And so some of what we're likely to be seeing in terms of a slower recovery of our business in those regions is confounded by the fact that distributors may be less -- more negatively impacted in their own business operations than NanoString would be in a direct market. So now I will just add that, I think that's probably the only color I can really add to that.

No, that's super helpful. And then just turning quickly to the SMI TAP program. I know you guys have said previously that it was meant to inform the development of the contact on the SMI and to refine the visual and analytics tools as well. Can you provide a brief update on how that development has progressing on these 2 fronts in the TAP program?

Yes. Happy to. Yes. So one of the things that the TAP allows us to do is see where the demand is from customers in terms of key parameters like the amount of area of a tissue that they want to analyze, the number of targets that they want to analyze, and whether the question they're asking is really more about building an atlas of cells, just a math or whether it's something that is actually trying to probe how any 2 individual cells are interacting. And so those are all very different use cases, each of which would inform a different part of the design choice at the instrument, the consumable and the software.

We have prototype software for presenting and visualizing spatial molecular imager data that we've been able to present to some of the early TAP customers with and get the real-time feedback on how they would explore their data. And while I don't want to characterize any particular insights, I can say it's incredibly valuable, and it's just the right time in our product development effort to continue to evolve the requirements. So we look forward to being in a position at a time in the future to demonstrate those tools more broadly.

Got it. Thank you for that. And then one final one for me. Stepping back, looking at a higher level, and given the recent emergence of participants in the spatial biology field, what are some of the key competitive advantages you guys are seeing at NanoString? I know you've previously highlighted the thousand Flex capability that is by and you earlier alluded to a higher throughput on the GeoMx. Are there any other key factors that have been jumping out? And secondly, is there anything to note in terms of traction for some of these platforms that launched earlier this year from your side?

Yes. Spatial biology is truly the next revolution in the field of biological research, and it's understandably attracting a lot of innovators. There's innovators from larger companies like us and then many start-ups along the way. I think across both the molecular imaging and the profiling categories, NanoString really leverages many of the same competitive advantages. One is, we've been in business serving researchers for over a decade. And I think we have a deep and intimate understanding of what's required by the market, particularly amongst translational researchers. 2, is we are -- we have, for years, built big, robust, highly automated instrument systems that you can place in a lab and they just work over and over again, day in and day out. Yes, that's a new skill that many smaller companies are building for the first time. 3 is, unlike some of the new entrants, we're actually participating in the market and engaging with hundreds of customers every day, every week. And so I think that makes us smarter about what the customer base needs.

Then of course, on top of those, there's a series of technical advantages we have. At NanoString, we've always built every platform to be compatible with Formalin-Fixed Paraffin-Embedded tissues first and then later come back to adapt some more, call it, esoteric sample types like fresh frozen. So I think we're always going to have an advantage in that sample type over our competitors. We've also embraced the idea of model mix. So we build all our spatial platforms to be capable of both RNA and protein at the same time, which is going to be another sustainable advantage.

And then finally, because we're pretty tuned in, I believe, to the translational market needs, we tend to build platforms that are high throughput and high plex. And those tend to also, by the way, be those with the strongest recurring consumable revenue straits because plex and throughput correlate with how much people spend on consumables. So we feel really great about our competitive advantages and our positioning in the marketplace, both on a technical and then maybe on a more kind of corporate capability level.

Our last question comes from the line of Douglas Schenkel.

Just, I guess, 2 or 3 cleanups. First, you -- I think I remember you guys talking about an expectation heading into Q2 that there wasn't going to be any improvement in lab access. I think the hope was it would be better than that. But I think that's what was factored into guidance. I just want to confirm based on what you talked about in your prepared remarks that it was probably a little bit better than you expected and kind of how you're treating that as we think ahead to the back half of the year?

Yes. Thanks for the question, Doug. I believe that our access to labs, meaning how our sales reps got into labs in the second quarter was about in line with our expectations. Most of our sales activity is actually still taking place over the phone at overseas. Many of these research institutions, although they're up and running and doing research, really aren't eager to have outside vendors walking their hallways. I know our sales reps meet their customers in coffee shops, off-campus or do work over the phone. So in that regard, I think we're right in line with what we expected. And by lab access, you really met more kind of the activity levels that are happening inside of the lab, using our technology, I'd say that was also more or less in line during the second quarter.

I think our nCounter consumable pull-through was about where we expected it to be during Q2. Our GeoMx consumable pull-through was better than we expected it to be during Q2. So I'd say that in that sense our lab access -- lab activity levels was in line or maybe slightly better than we had expected in Q2, with all the caveats about the sort of geographically uneven improvements that I mentioned earlier.

And I apologize if I missed this in your prepared remarks, but the 20 net placements or so for nCounter, was that sort of just one of those kind of timing dynamics in terms of why that number wasn't a little bit higher and not necessarily a reflection of something like there's a big drop-off in attachment rates with GeoMx or something like that. Is that just kind of one of those normal quarterly fluctuations we see from time to time?

Yes. It absolutely is. Doug, this is Tom. Yes. And if you look back at the first quarter compared to the second quarter, we had a big installed base leap in the first quarter. So you could see some anomalies and some of that installed pace that happens seasonally relative to the revenue, which tends to be a bit more smooth for nCounter. So another way of answering that, but we did sell more units than we installed this quarter as compared to the first year. So revenue has been very, very strong, and we feel really good about it headed into the second half of the year, as we mentioned in our prepared remarks.

And my last one is really on -- it's sort of a 2022 question. So I'll totally understand if you don't want to go there. But based on what you're seeing, I guess I'm just wondering how you're feeling about -- it's kind of related to what I just asked about, like, how you're feeling about the ability to keep placing nCounters, not just over the next 2 quarters, but over the next several quarters that are really influenced by the continued interest in GeoMx.

In my model, I actually have the pace of nCounter placements moderating next year in part because my assumption is that there's going to be more and more of these GeoMx instruments going in front of sequencers. Those things don't have to be mutually exclusive, right? You could place a lot of GeoMx instruments in front of sequencers and you could keep influencing and creating demands on the core nCounter side. Is it too early to say, look, your model is way too conservative based on what we're seeing? Or for now, do you think that this is a fair way of thinking about it, meaning GeoMx demand is going to continue to be robust, but the mix of placements is going to increasingly skew to basically in front of sequencers.

Yes. Great question, Doug. And I guess I'll start on GeoMx. I think you're correct that GeoMx placements will continue to skew towards NGS readout. That being said, I do not believe that negatively impacts the nCounter trajectory, because nCounter has a lot of uses well beyond GeoMx readout. So year-to-date, and actually, if you look back even to last year, about 20% of our GeoMx systems have been sold with an nCounter as bundles. And that percentage has been quite stable. So if as GeoMx grows, so too does -- if we remain at 20%, so too does the nCounter pull-through onto GeoMx. So that continues to be good.

On top of that, we've worked hard to diversify nCounter into areas like immunology and neurology, and I talked on the phone about stem cells, that has sustained nCounter instrument placements at 2019 rates really through the first half of this year. So truly, if you look pre-pandemic, kind of write-off 2020 and then look at 2021 post-pandemic, we're placing instruments at almost exactly the same pace. And so I don't see a reason that, that would moderate materially in 2022. I think it's conservative to model a little moderation, but really, there's not a trend I could point to that would make me expect it or worried about it.

And that concludes the Q&A session. I will now turn the call over back to Mr. Doug Farrell, Vice President of Investor Relations. Please go ahead with your closing remarks.

Thanks very much. Thank you, everyone, for joining us today. If you did miss any portion of the call, there will be a replay available for the next week. We'd ask callers to dial -- U.S. callers to dial 1-800-585-8367. International callers, please use (416) 621-4642. The conference ID is the same number for both. It is 3414117. So with that, we will say goodbye, and thank you for your time.

Thank you again for participating. This concludes today's conference call. You may now disconnect.